摘要
选取台风"菲特"(Fitow,201323)临近登陆过程为试验个例,在WRF模式基础上,采用4种不同的初始化方案,对台风"菲特"进行了72 h预报试验,并分析了模式的初始化对预报效果的影响。试验结果表明,对于台风路径的预报,使用集合平均作为初始场进行预报,预报结果相对直接使用GFS资料作为初始场进行预报的结果有明显改善,使用3DVAR同化方法,将HY-2A卫星散射计风场资料同化到集合平均的初始场中,台风路径预报进一步有所改善,而使用Hybrid同化方法将HY-2A卫星散射计风场资料同化到集合平均的初始场中,台风路径进而又有明显改善;但是在台风强度方面的预报,4种初始化方案效果不相上下。
Taken the Typhoon Fitow （201323） which is landing for example, four kinds of initialization schemes and 72-hour forecasts experiments are performed based on Weather Research and Forecasting （WRF） Model, and the influences of the initialization on the performance of the forecasts are analyzed. The results show that the performance of the track forecast using the ensemble-mean as the initial field is better than that using the GFS data. Assimilating the H^-2A scatter wind data into the initial field using 3DVAR assimilation method, the forecast of track is improved to some extent. Moreover, the performance of track forecast when ingesting the HY-2A scatter wind data into the initial field with Hybrid assimilation method is improved. However, the effect of the four kinds of method in the forecasts of intensity is roughly the same.
出处
《海洋预报》
2016年第1期1-10,共10页
Marine Forecasts
基金
国家海洋局海洋公益性行业专项(201305032-2)
